# Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Lists

arXiv: https://arxiv.org/abs/2203.09192 · HTML: https://mlsanity.com/paper/2203.09192/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 60 references that are in the map.
- First posted on arXiv 17 Mar 2022.

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [What Does BERT Look At? An Analysis of BERT's Attention](https://mlsanity.com/paper/1906.04341/) (2019)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [Attention is not not Explanation](https://mlsanity.com/paper/1908.04626/) (2019)
- [Generating Sequences With Recurrent Neural Networks](https://mlsanity.com/paper/1308.0850/) (2013)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [Quantifying the Carbon Emissions of Machine Learning](https://mlsanity.com/paper/1910.09700/) (2019)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification](https://mlsanity.com/paper/1903.04561/) (2019)
- [Is Attention Interpretable?](https://mlsanity.com/paper/1906.03731/) (2019)
- [Revealing the Dark Secrets of BERT](https://mlsanity.com/paper/1908.08593/) (2019)
- [Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases](https://mlsanity.com/paper/1909.03683/) (2019)
- [Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes](https://mlsanity.com/paper/1711.08412/) (2017)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [Adversarial Removal of Demographic Attributes from Text Data](https://mlsanity.com/paper/1808.06640/) (2018)
- [Reducing Gender Bias in Abusive Language Detection](https://mlsanity.com/paper/1808.07231/) (2018)
- [Multilingual and Multi-Aspect Hate Speech Analysis](https://mlsanity.com/paper/1908.11049/) (2019)
- [On Identifiability in Transformers](https://mlsanity.com/paper/1908.04211/) (2019)
- [Debiasing Pre-trained Contextualised Embeddings](https://mlsanity.com/paper/2101.09523/) (2021)
- [Challenges in Automated Debiasing for Toxic Language Detection](https://mlsanity.com/paper/2102.00086/) (2021)
- [Contextualizing Hate Speech Classifiers with Post-hoc Explanation](https://mlsanity.com/paper/2005.02439/) (2020)
- [Unmasking Contextual Stereotypes: Measuring and Mitigating BERT's Gender Bias](https://mlsanity.com/paper/2010.14534/) (2020)
- [Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Models](https://mlsanity.com/paper/2101.09688/) (2021)
- [What's in a Name? Reducing Bias in Bios without Access to Protected Attributes](https://mlsanity.com/paper/1904.05233/) (2019)
- [Demographics Should Not Be the Reason of Toxicity: Mitigating Discrimination in Text Classifications with Instance Weighting](https://mlsanity.com/paper/2004.14088/) (2020)

## Built on by (most cited first, 30 of 55)

- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models](https://mlsanity.com/paper/2308.01263/) (2023)
- [The Power of Noise: Redefining Retrieval for RAG Systems](https://mlsanity.com/paper/2401.14887/) (2024)
- [A Survey on Fairness in Large Language Models](https://mlsanity.com/paper/2308.10149/) (2023)
- [Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes](https://mlsanity.com/paper/2402.01981/) (2024)
- [Training-free Diffusion Model Adaptation for Variable-Sized Text-to-Image Synthesis](https://mlsanity.com/paper/2306.08645/) (2023)
- [Sociodemographic Bias in Language Models: A Survey and Forward Path](https://mlsanity.com/paper/2306.08158/) (2023)
- [Attention Entropy is a Key Factor: An Analysis of Parallel Context Encoding with Full-attention-based Pre-trained Language Models](https://mlsanity.com/paper/2412.16545/) (2024)
- [Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metric](https://mlsanity.com/paper/2402.06900/) (2024)
- [Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Survey](https://mlsanity.com/paper/2406.05392/) (2024)
- [Walking in Others' Shoes: How Perspective-Taking Guides Large Language Models in Reducing Toxicity and Bias](https://mlsanity.com/paper/2407.15366/) (2024)
- [Does Liking Yellow Imply Driving a School Bus? Semantic Leakage in Language Models](https://mlsanity.com/paper/2408.06518/) (2024)
- [Bias A-head? Analyzing Bias in Transformer-Based Language Model Attention Heads](https://mlsanity.com/paper/2311.10395/) (2023)
- [Towards Transformer-Based Aligned Generation with Self-Coherence Guidance](https://mlsanity.com/paper/2503.17675/) (2025)
- [Should We Attend More or Less? Modulating Attention for Fairness](https://mlsanity.com/paper/2305.13088/) (2023)
- [Tackling Bias in Pre-trained Language Models: Current Trends and Under-represented Societies](https://mlsanity.com/paper/2312.01509/) (2023)
- [Explaining Modern Gated-Linear RNNs via a Unified Implicit Attention Formulation](https://mlsanity.com/paper/2405.16504/) (2024)
- [Target-Agnostic Gender-Aware Contrastive Learning for Mitigating Bias in Multilingual Machine Translation](https://mlsanity.com/paper/2305.14016/) (2023)
- [Robust Natural Language Understanding with Residual Attention Debiasing](https://mlsanity.com/paper/2305.17627/) (2023)
- [Weigh Your Own Words: Improving Hate Speech Counter Narrative Generation via Attention Regularization](https://mlsanity.com/paper/2309.02311/) (2023)
- [The Lou Dataset -- Exploring the Impact of Gender-Fair Language in German Text Classification](https://mlsanity.com/paper/2409.17929/) (2024)
- [Feint and Attack: Attention-Based Strategies for Jailbreaking and Protecting LLMs](https://mlsanity.com/paper/2410.16327/) (2024)
- [Self-Explaining Hate Speech Detection with Moral Rationales](https://mlsanity.com/paper/2601.03481/) (2026)
- [Why Don't Prompt-Based Fairness Metrics Correlate?](https://mlsanity.com/paper/2406.05918/) (2024)
- [LLM Bias Detection and Mitigation through the Lens of Desired Distributions](https://mlsanity.com/paper/2510.06354/) (2025)
- [Progressive Feature Self-reinforcement for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2312.08916/) (2023)
- [The Life Cycle of Large Language Models: A Review of Biases in Education](https://mlsanity.com/paper/2407.11203/) (2024)
- [TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors](https://mlsanity.com/paper/2601.17958/) (2026)
- [Learning Hierarchical Relational Representations through Relational Convolutions](https://mlsanity.com/paper/2310.03240/) (2023)
- [Detecting Bias in Large Language Models: Fine-tuned KcBERT](https://mlsanity.com/paper/2403.10774/) (2024)

## Abstract (arXiv)

Natural Language Processing (NLP) models risk overfitting to specific terms in the training data, thereby reducing their performance, fairness, and generalizability. E.g., neural hate speech detection models are strongly influenced by identity terms like gay, or women, resulting in false positives, severe unintended bias, and lower performance. Most mitigation techniques use lists of identity terms or samples from the target domain during training. However, this approach requires a-priori knowledge and introduces further bias if important terms are neglected. Instead, we propose a knowledge-free Entropy-based Attention Regularization (EAR) to discourage overfitting to training-specific terms. An additional objective function penalizes tokens with low self-attention entropy. We fine-tune BERT via EAR: the resulting model matches or exceeds state-of-the-art performance for hate speech classification and bias metrics on three benchmark corpora in English and Italian. EAR also reveals overfitting terms, i.e., terms most likely to induce bias, to help identify their effect on the model, task, and predictions.